|Publication number||US7847846 B1|
|Application number||US 11/749,510|
|Publication date||Dec 7, 2010|
|Filing date||May 16, 2007|
|Priority date||May 16, 2006|
|Publication number||11749510, 749510, US 7847846 B1, US 7847846B1, US-B1-7847846, US7847846 B1, US7847846B1|
|Inventors||Zeljko Ignjatovic, Mark F. Bocko|
|Original Assignee||University Of Rochester|
|Export Citation||BiBTeX, EndNote, RefMan|
|Patent Citations (22), Non-Patent Citations (1), Referenced by (18), Classifications (8), Legal Events (2)|
|External Links: USPTO, USPTO Assignment, Espacenet|
The present application claims the benefit of U.S. Provisional Patent Application No. 60/800,420, filed May 16, 2006, whose disclosure is hereby incorporated by reference in its entirety into the present application.
The present invention is directed to CMOS image sensors and specifically to designs that employ readout transistor current sensing and feedback circuits to increase current driving capabilities and readout speed. The readout amplifier is reconfigurable and may be employed in both the reset and readout phases of the image sensor operation.
Two technologies are dominant in the field of image capture—the CCD and CMOS image sensors. Charge-coupled-device (CCD) imagers were first developed in the 1970's in the defense sector but quickly found acceptance among professional astronomers when they began using CCD cameras instead of photographic film. There were several advantages that favored CCD cameras over photographic film:
Throughout the last two decades, CCD's have been the most widely used imaging technology. They have continued to improve, and their principal advantages such as sensitivity (or quantum efficiency) of about 80%, very high dynamic range (up to five orders of magnitude) and linearity remain the highest among all commercially available image sensors. However, the biggest disadvantage of CCD imagers is their small format and/or high price due to the fact that CCD sensors rely on specialized fabrication processes that require dedicated and costly manufacturing lines. The drawbacks of CCD imaging technology are the following:
By the late 1990's CMOS image sensors began replacing CCD's in less demanding imaging applications. CMOS is the prevalent technology for digital circuit designs due to its favorable power-speed tradeoff and the high level of integration possible. As a result, CMOS has also attracted significant efforts toward implementation in a variety of mixed-signal designs. Thus, both analog preprocessing (filtering, amplification and A/D conversion) and digital processing have been combined in single-chip designs that may be fabricated with relatively inexpensive processes. CMOS technology also may be used to create inexpensive and effective image sensors. Although the sensitivity, dynamic range, and linearity of CMOS imagers are inferior to CCD imagers, CMOS imagers have other advantages. For example, CMOS image sensors can be made at standard manufacturing facilities that produce almost 90% of all semiconductor chips, which significantly decreases their design and manufacturing costs. Along with less expensive design and manufacturing, CMOS imagers have the following advantages:
Small pixel area (high spatial resolution), high dynamic range (equally good response to both low and high light intensities), high linearity, and high frame rate are the most important features required of CMOS imager designs. Recently, with the development of a variety of the battery-powered devices, low power consumption has also emerged as an important requirement. Thus pixel readout circuits have become a critical design element of CMOS imagers and a variety of approaches have been employed, which differ mostly by the number of pixels serviced by a single readout circuit. Designs range from readout circuits dedicated to each pixel up to circuits that are used to read out blocks of pixels, the various designs making different trade-offs in size, power, and performance.
The Active Pixel Sensor (APS) readout design proposed in E. R. Fossum, S. Mendis, S. E. Kemeny, “Active Pixel Sensor with Intra-Pixel Charge Transfer”, U.S. Pat. No. 5,471,515, November 1995, is used in nearly all commercial CMOS image sensors. The APS design places an amplifier at every pixel. In the APS design, the photodetector is isolated from the readout lines reducing crosstalk and charge loss, which was a significant drawback of previous Passive Pixel Sensor designs. The photodetector may be implemented as a photogate, as taught in E. R. Fossum, S. E. Kemeny, B. Pain, “Active Pixel Sensor with Multiresolution Readout”, U.S. Pat. No. 5,949,483, September 1999, or as a photodiode, as taught in E. R. Fossum, “Quantum Efficiency Improvements in Active Pixel Sensors”, U.S. Pat. No. 6,005,619, December 1999. Also, it could be implemented as a pinned photodiode, as taught in R. C. Gee, P. P. Lee, T. H. Lee, E. R. Fossum, “CMOS Active Pixel Sensor Using a Pinned Photo Diode”, U.S. Pat. No. 6,320,617 B1, November 2000, where the collected charge is transferred through a transfer gate to a floating diffusion area for charge-to-voltage conversion prior to the readout operation.
As shown in
Although APS designs have been able to realize many of the advantages associated with CMOS image sensors, process variation in the transistors located at the pixels cause relatively high DC offset and gain variation related fixed pattern noise (FPN) as opposed to CCD's that intrinsically have low FPN due to the absence of active devices at the pixels. Furthermore, the minimum size readout transistors used in APS pixel designs introduce excess 1/f noise and thermal channel noise thereby reducing the dynamic range of CMOS imagers, which remains less than the dynamic range of CCD imagers. In order to reduce the DC offset related FPN and 1/f noise of the readout transistors correlated double sampling (CDS) techniques have been proposed. In the conventional CDS method the terminal voltage of the photo-detector is sampled twice; once after the reset phase and once after the integration phase. The two voltage values are subtracted in an external CDS circuit and the resulting difference voltage corresponds to the incident light intensity (to first order). Thus, the DC offset noise is cancelled and the 1/f noise is differentiated, which greatly attenuates it due to its slowly varying nature.
There are two CDS approaches that differ by the time instances at which the reset and readout samples are taken. If CDS is performed such that the voltage corresponding to the end of the reset phase is sampled first following by the voltage sample after the integration phase, the DC offset related FPN and reset noise are removed. However, even though the 1/f noise is differentiated, it is poorly attenuated because the 1/f noise samples are weakly correlated due to the relatively large time interval between samples. Another disadvantage of this CDS approach is that it requires a separate capacitive storage element for each pixel. The capacitive element may be placed at the pixel site as proposed in R. M. Guidash, “Three Transistor Active Pixel Sensor Architecture with Correlated Double Sampling”, U.S. Pat. No. 6,587,146, July 2003, and R. B. Merrill, “Active Pixel Sensor with Noise Cancellation”, U.S. Pat. No. 6,940,551 B2, September 2005; however, these approaches require that additional transistors be placed at the pixel site which reduces the pixel area that otherwise might be used for light gathering. The capacitive elements could be placed external to the pixel area; however, this approach may be prohibitively expensive in terms of silicon area and it has been shown that the CDS circuits may require as much area as the imaging region itself, see
The other CDS approach reverses the sampling procedure of the previous method. In this second method the voltage at the end of the light integration phase is sampled first followed by sampling the photo-detector voltage after the reset operation. These two values are subtracted in CDS circuits to remove the DC offset and attenuate the 1/f noise. The time difference between the sampled voltage values is now much shorter than in the first CDS method, so that the 1/f noise samples are more highly correlated and thus more greatly attenuated by differentiation. This CDS method requires only one capacitive element per column, which consumes much less silicon area than the previous CDS method. However, this second method doubles the reset noise contribution which becomes more prominent with technology scaling, i.e., as the photodiode capacitance, C, decreases the reset voltage noise power spectral density, kT/C, increases.
Another important disadvantage of conventional APS image sensors, which are based on a source-follower amplifier at each pixel site, is the relatively large gain error among the pixels of the imager. As a result, there is relatively large gain related FPN. As stated above, the APS design is a source-follower structure that has a voltage closed-loop gain given by Eq. (1),
where gm is the transconductance of Q2 (see
In order to improve the gain related FPN problem arising in APS designs, the Active Column Sensor (ACS) design was proposed in M. A. Pace, J. J. Zarnowski, “Complementary Metal Oxide Semiconductor Imaging device”, U.S. Pat. No. 6,084,229, July 2000. The ACS schematic is shown in
The pixel unit in the ACS design contains three transistors. The reset transistor Q1 106 connects the photo-element to the reset voltage during the reset operation. Transistor Q2 108 serves as a switch to multiplex the in-pixel transistor Q3 112 to the common amplifier circuit shared among the pixels of one row. The remainder of the amplifier circuit is placed outside of the pixel area. The in-pixel amplification transistor Q3 112 together with the external transistors Q4 302 and Q5 304 constitute an input differential pair of a unity gain differential amplifier (UGA). The transistor Q8 306 provides current biasing to the differential pair. The current mirror Q6 308 and Q7 310 provides an active load to the differential pair. The output of the amplifier is tied to the negative input providing negative feedback, thus creating a unity gain buffer circuit topology. The main advantage of the ACS design over the APS design is that the open-loop gain can be increased significantly without increasing the size of the in-pixel transistors. This may be achieved by placing cascode transistors outside the pixel area to boost the open-loop gain in the UGA structure. By increasing the open-loop gain, device parameter variation does not affect the closed-loop gain associated with each pixel as much as in the APS design. Consequently, significantly lower gain related FPN has been reported for the ACS design.
Correlated double sampling may be employed in ACS designs to remove the dc offset, 1/f noise and reset noise. Similar to APS, there are two CDS methods that may be employed with the ACS approach. During the reset phase, both switches Q1 and Q2 are turned on. The transistor Q1 connects the photo-element to the reset voltage value Vreset. The transistor Q3 is connected to the amplifier's body through the bus lines. As a result, a unity-gain differential amplifier UGA is formed and it outputs the value Vout
An objective of the present invention is to combine the small pixel size and low power consumption of the conventional APS design with faster settling times and correspondingly increased readout speeds. A further objective is to reduce reset noise by utilizing the in-pixel amplifier in a negative feedback configuration.
To achieve the above and other objectives, the present invention is directed to a current sensing active pixel (CSAP) sensor design in which the current through the in-pixel readout transistor is sensed by a circuit that is external to the pixel and according to the measured current value a feedback current is supplied to charge the read-line parasitic capacitance. The feedback current is supplied by a circuit that also is external to the pixel area. The amplifier structure is reconfigurable so that it can be used both to read out and to reset the pixel. The in-pixel readout transistor is in a source-follower configuration during the pixel readout operation, and during the pixel reset operation it is a common-source type amplifier, similar to the method proposed in the following: L. J. Kozlowski, D. L. Standley, “Low-Noise Active Pixel Sensor for Imaging Arrays with Global Reset”, U.S. Pat. No. 6,493,030, December 2002; L. J. Kozlowski, D. L. Standley, “Compact Low-Noise Active Pixel Sensor with Progressive Row Reset”, U.S. Pat. No. 6,697,111 B1, February 2004; and L. J. Kozlowski, D. L. Standley, “Active Pixel Sensor with Capacitorless Correlated Double Sampling”, U.S. Pat. No. 6,535,247, March 2003. During both the readout and reset operations the in-pixel amplifier operates in conjunction with the external circuit that provides current sensing and feedback.
The present invention allows a new CMOS image sensor readout that is several hundred times faster than standard CMOS active pixel sensor (APS) designs. In addition to the obvious advantages of higher frame rates for large pixel arrays and faster electronic shutter, increasing the readout speed reduces read noise by decreasing the time interval between the successive measurements in correlated double sampling. In the new method the in-pixel readout transistor is biased by a constant current source shared by the pixels of a row/column. The transistor is used in the source-follower configuration however, in contrast to standard APS designs the biasing current through the readout transistor is monitored and negative feedback forces the current to a fixed value. The external voltage required to maintain the fixed current is then sensed.
The present invention retains the high fill factor of APS designs by employing only three transistors at the pixel site. The transistor current sensing and feedback circuit is external to the pixel so its transistors may be sized to provide much higher current driving capability thereby enabling readout speeds that are several hundred times faster than standard APS designs. Due to its high speed, the present invention can more effectively employ correlated double sampling (CDS) significantly reducing the read noise contribution of the 1/f noise of the source follower transistor.
The present invention also reduces gain variation related fixed pattern noise (FPN) in comparison to conventional APS designs by increasing the effective open-loop gain of the source-follower amplifier. Finally, the new method greatly reduces reset noise by placing the reset transistor in the feedback path of an inverting amplifier which actively drives the photodiode reset voltage to a reference voltage level. The readout amplifier is reversible and therefore may be used in both the reset and the readout phases. The present invention also may be employed in multiplexed configurations in which neighboring photodetectors share a common amplification transistor thereby enabling designs with 1.5 transistors per pixel while still performing the aforementioned functions.
Preferred embodiments of the present invention will be disclosed in detail with reference to the drawings, in which:
Preferred embodiments of the present invention will be set forth in detail with reference to the drawings, in which like reference numerals refer to like elements throughout.
A general CSAP architecture configured for pixel readout operation is shown in
The negative feedback operation provided by the CSAP structure is described in the following. If the input voltage value at the gate of Q3 increases, the gate-source voltage VGS3 increases and the current in Q3 increases to ID3, which is now larger than the biasing current Ibias. The sensing circuit senses this current increase and outputs a current that is equal to ID3. The feedback circuit compares Lsens and Ibias multiplies their difference, and supplies the ROW_BUS2 line with the additional current Io=Ai*(Isens−Ibias)=Ai*(ID3−Ibias). This additional current is positive, because Isens>Ibias; hence, it charges the parasitic capacitance of the ROW_BUS2 line. As a result, the potential of the ROW_BUS2 line (which is at the same time the source terminal of Q3) increases to reduce the gate-source voltage of Q3 thereby reducing ID3 to a value that is equal to Ibias. Similarly, if the input voltage value at the gate of Q3 decreases, the additional current Io becomes negative increasing the total current supplied to the ROW-BUS2 line to discharge the parasitic capacitance. As a result, the potential of the ROW_BUS2 line and source terminal of the Q3 decreases giving rise to the gate-source voltage VGS3 and the subsequent increase of ID3 to a value that is equal to Ibias. The transfer function and −3 dB frequency, which determines the circuit speed, of the CSAP architecture during the readout operation are given by Eq. (2) and (3), respectively. Hence, the −3 dB frequency is increased by a factor that is proportional to the current gain Ai(kAi) with respect to standard APS. During the readout operation the reset transistor Q1 is turned off.
In another embodiment of this invention, the potential of the ROW_BUS1 line (VX) could be set with an additional voltage level-shifter circuit such that it is held above the potential of the ROW_BUS2 line (VY) by a fixed voltage value of ΔV (VX=VY+ΔV). As a result, the drain-source voltage (Vds3 3) of the in-pixel amplification transistor is held to a fixed value of ΔV. Since the Vds3 is fixed during the readout operation, the influence of the channel length modulation (finite output resistance rds3) may be neglected. Therefore, since the drain current ID3 through the amplification transistor is independent on drain-source voltage, the large-signal response of the CSAP with fixed Vds 3 exhibits highly linear response. This advantage of the CSAP with fixed Vds 3 in terms of linearity is even more emphasized with technology scaling, where the channel length modulation and other second-order effects such as drain-induced barrier lowering (DIBL) may cause a high dependence of the drain current on the drain-source voltage resulting in a non-linear response of the standard APS. A general CSAP architecture with fixed Vds 3 is shown in
It has been shown that the in-pixel transistor (Q3) could be reconfigured in a common-source amplifier architecture in order to attenuate the thermal (reset) noise injected by the reset switch into the photodetector element. In this active reset configuration, the reset switch is placed in the negative feedback of the amplifier. Ideally, assuming that the common-source amplifier has an infinite bandwidth, the total reset noise would be reduced by a factor of
where, ADC is the open-loop DC gain of the common-source amplifier (ADC=gm3*rds3), Ron and rds3 are the on-resistances of the reset switch Q1 and amplification transistor Q3, respectively. However, the pixel-level common-source amplifier is heavily loaded by the parasitic capacitance CX of the readout lines, which significantly reduces its unity-gain bandwidth. The unity-gain bandwidth of such an amplifier is readily lower than the thermal noise cut-off frequency, so that the most of the thermal noise power is still injected into the photodetector without any attenuation. As a result, due to the reduced bandwidth of the in-pixel amplifier, the circuit reset noise attenuation capabilities are limited and only a modest noise reduction by a factor of 2 has been reported. Our current sensing method can be used to extend the bandwidth of the common-source amplifier reducing the reset noise that is injected by the reset switch Q1. The general CSAP architecture configured for the pixel reset operation is shown in
The negative feedback provided by the CSAP structure during the reset operation may be described as follows. If the input voltage value at the gate of Q3 increases, the gate-source voltage VGS 3 increases and Q3 starts to conduct more current ID3, which is now larger than the biasing current Ibias. The sensing circuit senses this current increase and generates a current that is equal (or proportional with constant k) to ID3 at its output. The feedback circuit compares Isens and Ibias (or their k times scaled versions), multiplies their difference, and supplies the ROW_BUS1 line with the additional current Io=Ai*(Ibias−Isens)=Ai*(Ibias=ID3). This additional current is negative, because Isens>Ibias, and is subtracted from the bias current Ibias. As a result, the supply current on the ROW_BUS1 line is reduced which causes the potential of the ROW_BUS1 line (which is also the potential of the drain terminal of Q3) to decrease. The voltage decrease at the drain of Q3 is fed back through the switch Q1 to the gate of Q3 decreasing its initial rise. Similarly, if the input voltage value at the gate of Q3 decreases ID3 3 decreases and the additional current Io becomes positive. This result in an increase of the total current supplied to the ROW-BUS I line, so that the potential of the ROW_BUS1 line and drain terminal of Q3 increases. This voltage increase gives rise to the gate-source voltage VGS 3 and increases ID3 to a value that is equal to the Ibias. The transfer function of the equivalent common-source amplifier during the reset operation and its unity-gain frequency fta are given by Eq. (4) and (5), respectively. Since the unity-gain frequency of the amplifier proposed in the above-referenced Yang et al paper equals to gm3/(2πCX), the feedback operation of the CSAP architecture increases the unity-gain frequency and bandwidth of the common-source amplifier by a factor of Ai(kAi). Therefore, more reset noise power will fall into the amplifier's bandwidth and will be attenuated by the negative feedback of the common-source amplifier. Assuming that the cut-off frequency of the thermal noise is lower than the unity-gain frequency given by Eq. (5), the total reset noise would be attenuated by the amplifiers open-loop DC gain (gm3*rds3)*(1+rds3/rds1).
In another embodiment of the present invention, the amplification transistor Q3 and access transistor Q2 are interchanged such that the access transistor connects/disconnects the source of the Q3 to/from the ROW_BUS2 line instead of ROW_BUS1 line.
Similarly to the method proposed in Yang et al, the reset noise injected into the photo-detector could be further suppressed by using the access transistor Q2 as a cascode device (saturation mode) rather than a switch (triode region). This cascode type common-source amplifier configuration is achieved by applying an appropriate voltage bias on the CS line such that both transistors Q2 and Q3 are always in saturation region during the reset operation. The resulting open-loop transfer function of the common-source amplifier is given by Eq. (6). The open-loop DC gain is increased gm2*rds2 times with respect to the configuration where the access transistor Q2 is biased in triode mode. The unity-gain frequency of this cascode type common-source configuration is equal to the unity-gain frequency of the configuration where the access transistor is biased in triode region, shown in Eq. (5). We conclude that if the access transistor is biased in saturation region, the reset noise reduction is further improved. This improvement results from the increased open-loop DC gain of the amplifier provided that the current sensing circuitry maintains fairly high circuit speed such that most of the reset noise power lays within the circuit bandwidth.
The current sensing circuit shown in
However, in the implementations shown in
The proposed CSAP design was simulated using a resistive type of current sensing circuit, the schematic of which is shown in
The parasitic capacitance of C2 of the bottom ROW_BUS line is the dominant parameter over Cl of the top ROW_BUS line. Changing C2 will affect the response time and readout speed of the proposed design effectively, such as increasing C2 X times with C1 fixed will results in X times longer settling times.
CSAP with A = 10
CSAP with A = 20
CSAP with A = 40
While a preferred embodiment of the present invention and various modifications thereof have been set forth above, those skilled in the art who have reviewed the present disclosure will readily appreciate that other embodiments can be realized within the scope of the invention. The photo-detectors that are shown in the preferred embodiments of this invention could be implemented as photodiodes (three-transistor 3T CSAP readout structure), photogates with transfer gate (4T CSAP readout structure with photogates), and pinned diodes with floating diffusions and transfer gate (4T CSAP readout structure with pinned diode). Therefore, the invention should be construed as limited only by the appended claims.
|Cited Patent||Filing date||Publication date||Applicant||Title|
|US5471515||Jan 28, 1994||Nov 28, 1995||California Institute Of Technology||Active pixel sensor with intra-pixel charge transfer|
|US5949483||Jan 22, 1997||Sep 7, 1999||California Institute Of Technology||Active pixel sensor array with multiresolution readout|
|US6005619||Oct 6, 1997||Dec 21, 1999||Photobit Corporation||Quantum efficiency improvements in active pixel sensors|
|US6084229||Mar 16, 1998||Jul 4, 2000||Photon Vision Systems, Llc||Complimentary metal oxide semiconductor imaging device|
|US6252217 *||Dec 7, 1998||Jun 26, 2001||Simage Oy||Device for imaging radiation|
|US6320617||Oct 31, 1996||Nov 20, 2001||Eastman Kodak Company||CMOS active pixel sensor using a pinned photo diode|
|US6493030||Apr 8, 1998||Dec 10, 2002||Pictos Technologies, Inc.||Low-noise active pixel sensor for imaging arrays with global reset|
|US6535247||May 19, 1998||Mar 18, 2003||Pictos Technologies, Inc.||Active pixel sensor with capacitorless correlated double sampling|
|US6587146||Nov 20, 1998||Jul 1, 2003||Eastman Kodak Company||Three transistor active pixel sensor architecture with correlated double sampling|
|US6697111||Apr 8, 1998||Feb 24, 2004||Ess Technology, Inc.||Compact low-noise active pixel sensor with progressive row reset|
|US6777660 *||Feb 4, 2002||Aug 17, 2004||Smal Technologies||CMOS active pixel with reset noise reduction|
|US6940551||Sep 24, 2001||Sep 6, 2005||Foveon, Inc.||Active pixel sensor with noise cancellation|
|US7023369 *||Jul 14, 2004||Apr 4, 2006||University Of Rochester||Multiplexed-input-separated sigma-delta analog-to-digital converter design for pixel-level analog-to-digital conversion|
|US7145123 *||Aug 12, 2004||Dec 5, 2006||Stmicroelectronics S.A.||CMOS light sensing cell|
|US7215369 *||Apr 2, 2003||May 8, 2007||Micron Technology, Inc.||Compact pixel reset circuits using reversed current readout|
|US7282685 *||Apr 14, 2005||Oct 16, 2007||Micron Technology, Inc.||Multi-point correlated sampling for image sensors|
|US7466255 *||Nov 17, 2006||Dec 16, 2008||University Of Rochester||Multiplexed-input-separated Σ-Δ analog-to-digital converter for pixel-level analog-to-digital conversion utilizing a feedback DAC separation|
|US20050088554 *||Oct 26, 2004||Apr 28, 2005||John Scott-Thomas||Electrical sensing apparatus and method utilizing an array of transducer elements|
|US20080259182 *||Oct 20, 2005||Oct 23, 2008||Simon Fraser University||High Gain Digital Imaging System|
|US20080309800 *||Jun 18, 2007||Dec 18, 2008||Micron Technology, Inc.||Noise and parasitic capacitance reduction for 1t pixel cmos image sensors|
|US20090002533 *||May 12, 2008||Jan 1, 2009||Chen Oscal T-C||Apparatus of high dynamic-range CMOS image sensor and method thereof|
|US20090002536 *||Jul 3, 2008||Jan 1, 2009||Junichi Nakamura||Biasing scheme for large format CMOS active pixel sensors|
|1||Yang, J. et al., A 3MPixdel Low-Noise Flexible Architecture CMOS Image Sensor, 2006 IEEE International Solid-State Circuits Conference, Session 27, Feb. 8, 2006.|
|Citing Patent||Filing date||Publication date||Applicant||Title|
|US7999864 *||Jun 25, 2008||Aug 16, 2011||Canon Kabushiki Kaisha||Imaging apparatus and its control method for setting suitable clipping level|
|US8169526 *||Aug 28, 2009||May 1, 2012||Naoki Ozawa||Low noise signal reproducing method for a solid state imaging device|
|US8564369 *||Dec 6, 2011||Oct 22, 2013||Samsung Electro-Mechanics Co., Ltd.||Linear amplifier|
|US9052497||Mar 12, 2012||Jun 9, 2015||King Abdulaziz City For Science And Technology||Computing imaging data using intensity correlation interferometry|
|US9099214||Dec 11, 2012||Aug 4, 2015||King Abdulaziz City For Science And Technology||Controlling microparticles through a light field having controllable intensity and periodicity of maxima thereof|
|US9118316||Mar 26, 2012||Aug 25, 2015||Semtech Corporation||Low voltage multi-stage interleaver systems, apparatus and methods|
|US9200954||Nov 7, 2012||Dec 1, 2015||The Johns Hopkins University||Flexible readout and signal processing in a computational sensor array|
|US9344655||Jun 29, 2015||May 17, 2016||Imec Vzw||Active pixel sensor imaging system|
|US9426393 *||Feb 11, 2014||Aug 23, 2016||Taiwan Semiconductor Manufacturing Co., Ltd.||Noise simulation flow for low noise CMOS image sensor design|
|US9571774 *||Sep 10, 2015||Feb 14, 2017||Analog Value Ltd.||Current mode pixel and readout circuit having a first and a second readout phase|
|US20090015696 *||Jun 25, 2008||Jan 15, 2009||Canon Kabushiki Kaisha||Imaging apparatus and its control method|
|US20110050970 *||Aug 28, 2009||Mar 3, 2011||Naoki Ozawa||Low noise signal reproducing method for a solid state imaging device|
|US20130208154 *||Feb 14, 2012||Aug 15, 2013||Weng Lyang Wang||High-sensitivity CMOS image sensors|
|US20150116506 *||Feb 11, 2014||Apr 30, 2015||Taiwan Semiconductor Manufacturing Co., Ltd.||Noise simulation flow for low noise cmos image sensor design|
|US20160240572 *||Sep 10, 2015||Aug 18, 2016||Vladimir Koifman||Current mode pixel and readout circuit|
|CN104380638A *||Mar 25, 2013||Feb 25, 2015||Sem技术公司||Low voltage multi-stage interleaver systems, apparatus and methods|
|EP2963815A1||Jun 30, 2014||Jan 6, 2016||IMEC vzw||Active pixel sensor imaging system|
|WO2013148578A1 *||Mar 25, 2013||Oct 3, 2013||Semtech Corporation||Low voltage multi-stage interleaver systems, apparatus and methods|
|U.S. Classification||348/301, 348/308, 250/208.1, 348/302|
|Cooperative Classification||H04N5/378, H04N5/3745|
|Sep 12, 2007||AS||Assignment|
Free format text: ASSIGNMENT OF ASSIGNORS INTEREST;ASSIGNORS:IGNJATOVIC, ZELJKO;BOCKO, MARK F.;SIGNING DATES FROM 20070625 TO 20070809;REEL/FRAME:019841/0815
Owner name: UNIVERSITY OF ROCHESTER, NEW YORK
|Jun 9, 2014||FPAY||Fee payment|
Year of fee payment: 4